Scientists just discovered a key connection between face processing and intelligence

New research has revealed that the ability to process faces is not only connected to general intelligence but is also negatively associated with certain traits linked to autism. The study, published in the Quarterly Journal of Experimental Psychology, found that individuals who performed well on face processing tasks also tended to score higher on intelligence tests. Conversely, those who reported more difficulties with nonverbal communication, a trait sometimes associated with autism, showed poorer face processing skills.

“Faces are such a critical part of our day-to-day lives, and most people don’t think about their face perception abilities,” said study author Dana L. Walker of the University of Western Australia. “There is a lot of information we extract from the face, such as someone’s identity, emotional state, trustworthiness, and so on. But research is still needed to fully understand how we perceive faces, and especially how our ability to perceive faces relates to other aspects of our personality and cognition.”

For their study, the researchers recruited 253 participants from the general community through the TestableMinds platform. To ensure a relatively homogenous sample and minimize confounding variables, they established specific inclusion criteria. Participants had to be between 18 and 49 years old to ensure cognitive abilities are relatively stable. Residence in Australia, USA, UK, New Zealand, Canada, or Ireland was required due to shared linguistic and cultural similarities. The researchers also required participants to have English as their first language.

Upon recruitment, the participants completed a battery of tests designed to measure different aspects of face processing abilities, cognitive abilities, and trait autism. Four tests were administered to assess different components of face processing. The Mooney Face Task measured face detection ability. The Cambridge Face Perception Test (CFPT) assessed the ability to discriminate or individualize faces from each other. The Cambridge Face Memory Test (CFMT) measured the ability to encode a face to memory and recall that face at a later time. The emotion labelling task evaluated the ability to accurately identify facial expressions related to emotion. The Cambridge Car Memory Test (CCMT) was also administered to serve as a non-face comparison for the CFMT.

To measure the cognitive abilities of the participants, the researchers employed four tests designed to assess different aspects of intelligence. The Concrete and Abstract Words Synonym Test measured crystallized intelligence, a measure of accumulated knowledge and verbal skills. The paper folding test assessed visuospatial ability, involving the ability to visualize and mentally manipulate shapes and objects. The visual working memory test measured an individual’s memory span, the ability to hold and manipulate information in their mind. The Baddeley’s Semantic Verification Test measured speeded reasoning, specifically the ability to quickly and accurately process information.

To assess traits associated with autism, the researchers used the Autism Spectrum Quotient (AQ), a questionnaire to measure the dimensions of autism: deficits in communication, poor social skills, and attention to detail.

The researchers confirmed the existence of a general face processing ability factor (f). This means that individuals who perform well on one type of face processing task (such as recognizing emotions) tend to perform well on other face processing tasks as well (such as remembering faces). This factor suggests that there is an underlying general ability related to how our brains process faces.

Secondly, the study found a moderate, positive association between this general face processing ability factor (f) and general intelligence (g). This suggests that individuals with higher cognitive abilities tend to be better at processing faces. This association lends support to the idea that face processing could be considered a cognitive ability that is related to general intelligence.

Thirdly, and perhaps most notably, the study revealed that difficulties in nonverbal communication, a trait often associated with autism, were negatively associated with the general face processing ability factor (f). This indicates that individuals who struggle with nonverbal communication tend to have poorer face processing abilities. This negative association remained even when the researchers took overall intelligence into account, which suggests that these communication difficulties may be an independent factor affecting how people process faces.

Finally, face perception and face memory were found to uniquely predict expression recognition ability. It was found that face detection, face perception, and face memory have a relationship where face detection predicted face perception, face perception predicted face memory, and face memory predicted expression recognition ability.

“Not everyone finds it easy to extract information from faces about someone’s identity and emotional state,” Walker told PsyPost. “There are varying degrees of face perception ability ranging from poor perception of faces to very good perception of faces, and these differences are associated with other cognitive abilities and aspects of our personality. By better understanding how these abilities and traits relate to each other, we may be able to develop interventions for those who struggle with face perception.”

The findings provide support for the idea that face processing skills should be considered a broad cognitive ability in their own right, similar to skills such as reasoning or spatial thinking. This opens up the possibility that models of intelligence, which have traditionally focused on areas like memory, speed, and reasoning, might be expanded to include social and perceptual skills such as face processing. The researchers even propose that deficits in face memory—often used as a basis for diagnosing a condition called developmental prosopagnosia—might be viewed through the lens of a learning disability.

Importantly, the study was preregistered. In other words, the researchers publicly documented the study’s design, hypotheses, and analysis plan before data collection. This practice helps increase transparency and reduces the risk of researchers unconsciously biasing their findings or selectively reporting results.

“Ongoing, transparent, and rigorous research is important,” Walker explained. “It can help redefine and refine important theoretical and practical areas of psychology. For example, the definition and theoretical structure of intelligence, which has implication for many areas of our lives (such as our education systems), is continuously evolving as more research is conducted. Also, the way we evaluate the diagnosis protocol of clinical prosopagnosia (the inability to recognise faces) needs to be evaluated and refine as we understand more about face recognition.”

But the study, like all research, includes limitations.

“One caveat is the generalisability of our research,” Walker noted. “We collect data from predominantly Caucasian individuals from Western countries. Therefore, our findings may or may not generalise to different ethnicities and cultures. More research would be needed to replicate our findings in other countries and with different races.”

“There is more we need to understand about face perception, intelligence, and trait-autism. As mentioned, it would be good to replicate these findings in different populations to see how generalizable our findings are. We hope that this research provides the foundation for future research into face perception, intelligence, and trait-autism, especially in the field of developmental psychology.”

The study, “The Inter-Association Between Face Processing, Intelligence, and Autistic-like Nonverbal Communication,” was authored by Dana Walker, Romina Palermo, and Gilles G. Gignac.